CFP For The VAS

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi All,

Most of the circuits I have seen here in this forum and elsewhere
utilize a single transistor for the VAS. Mr.Self, describing his
'Blameless Amplifier', maintains that it is best to use a darlington
connected pair for the VAS.I would like to hear your opinions
on swapping this darlington configuration with a Sziklai pair.

One downside I can think of is, in a darlington connection as in
Mr. Self's amps, the input transistor of the darlington pair
operates in an almost constant Vce because the collector is
tied to the ground.In a CFP configuration, both transistors
making up the VAS will see the rail-to-rail excursion, forcing
the first transistor to be a high voltage type as well.

Selim
 
Sziklai has advantages for voltage followers but the VAS is just a current amplifier, so using a darlington or a CFP is about the same.

View the input stage as a current source (it converts the error voltage into a current) if you have problems understanding this.

If you don't use a current mirror in the input stage, it's a bit less clear, though...
 
Thanks Sebastien. I have used this precise VAS configuration in one of my amp designs.

Make each stage as linear as possible, then apply feedback, not much at all in this case..... there is a LOT of degeneration in there.

Output stage, another CFP, shows signs of very great design care. I assume it's Class AB, however, with some issues at switching.

There is nothing new under the sun.

Thank you for sharing this circuit, Sebastien,

Cheers,

Hugh
 
Hi

It depends on how you perceive the circuit. The best VAS I have ever built consists of a bipolar, constant power, differential bridge. If you look at the constant power part, it resembles a CFP that happens to be cascode.:) So I guess my vote goes towards the current feedback in that sense.:up:
 
AKSA said:
No, it is not a misquote. Andrew is very precise in his contributions.

It is a reference to both transistors being identical types, that is, both npn (fixed at the negative rail) or both pnp (fixed at the positive rail).

Cheers,

Hugh


Wow! Never heard that kind of expression before. I thought he was joking or something... Well, another day I learned something new...


Thanks Hugh.
 
Here is an example of a CFP-Cascode VAS.
It was proposed by B.J. Codd in Wireless World, October 1979, pp 81-85.
Initial circuit :
 

Attachments

  • codd_preamp.jpg
    codd_preamp.jpg
    71.5 KB · Views: 1,407
CBS240 said:
Hi

It depends on how you perceive the circuit. The best VAS I have ever built consists of a bipolar, constant power, differential bridge. If you look at the constant power part, it resembles a CFP that happens to be cascode.:) So I guess my vote goes towards the current feedback in that sense.:up:


CBS240 said:
Hi forr

Now double it up to make a differential VAS, and don't skimp on the parts. Go all out for quality matched arrays.:yes:


Hi CBS
Well, I was going to ask you to show your best VAS...
I am very interested by constant power schemes.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.